Concerning assets, the impact of the drought is restricted to middle-assets households.

The impact of the drought on consumption is restricted to the poorest households in both countries.

In Malawi, the richest households successfully mitigate the impact of the drought by relying on agricultural-related coping strategies.

In Ethiopia, the richest households increase their time devoted to non-agricultural activities (sending a member to migrate for work, for example), and thanks to the higher labor productivity in this sector, they increase their real income.

The increase in income of the highest-income households in Ethiopia is explained by a reallocation of labor from the farm sector to the non-farm sector, which is not observed in Malawi.

My results show that the droughts significantly increase inequality in both countries. Low-income households experienced a drop in real income of 40%, while high-income households increased their income in Ethiopia and were not affected in Malawi.

I apply this method to three waves of geo-referenced household panel data from LSMS-ISA linked with a high-resolution drought index: the SPEI. I exploit the specific spatial and temporal pattern of droughts in the countries to isolate the causal impact of these droughts on income distribution.

In this paper, I measure the causal impact of the droughts of 2015 in Ethiopia and 2016 in Malawi on the distribution of income using a recent empirical method that generalizes the difference-in-differences method to the entire income distribution.

In this paper, I measure the causal impact of the droughts of 2015 in Ethiopia and 2016 in Malawi on the distribution of income using a recent empirical method that generalizes the difference-in-differences method to the entire income distribution.